[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference csc32::consolemanager

Title:POLYCENTER Console Manager
Notice:Kits, Scans, Docs on CSC32:: as PCM$KITS:,PCM$DOCS:, PCM$SCANS:
Moderator:CSC32::BUTTERWORTH
Created:Thu Aug 06 1992
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:1541
Total number of notes:6564

934.0. "unusual request" by 30188::CLEMENCE () Sat Aug 19 1995 01:02

One of the users of the PCM here gets annoyed by the disk doing an access
for everything thats goes in or out on a console port. 

What this person asked me to do was to buffer this disk access in memory for a
while and then burst out a clump of data to the log files; therby reducing the
amount of times thast the local vms disk gets accessed.

Since I don't know of any way to adjust PCM to buffer its data before writing
it to the log files; I said I would ask here in this notes conference on how
to do it.

If this is not possible, thats an ok answer too.


			Thanks
			Bill
T.RTitleUserPersonal
Name		DateLines
934.1No can doZENDIA::DBIGELOWInnovate, Integrate, EvaporateMon Aug 21 1995 10:547
Bill,

   There's nothing that can be done to satisfy your needs at the moment. If you
want, you can send a request to the product manager, MUZICK::KUNG, to have this 
implemented in the next release.

Dave
934.2CSC32::BUTTERWORTHGun Control is a steady hand.Mon Aug 21 1995 14:0913
    >One of the users of the PCM here gets annoyed by the disk doing an
    >access for everything thats goes in or out on a console port.
    
    I think he might be more annoyed if the PCM system crashed before it
    got a chance to flush the buffered data to disk. I'm not trying to be
    nasty here but there is a reason for the way it works like it does and
    I'm not saying it can't be changed in some way such as waiting until 
    we see a record terminator or say 5 seconds has elapsed before we
    write the buffer (We need the latter for those systems that do not send
    terminators).
    
    Regs,
      Dan
934.3I'm happy with the answer I got.30188::CLEMENCETue Aug 22 1995 09:0820
    
>    I think he might be more annoyed if the PCM system crashed before it
>    got a chance to flush the buffered data to disk. I'm not trying to be

I know that is a problem; for the person requesting they think that the little
loss of data is acceptable. I myself would prefer to get all the information
up to the last moment. 

	I have been though the times of not having that little bit of critical 
data and somehow I survived, but it was painfull!!! 


	 I was asked to see "what can be done" to reduce the amount of disk 
accesses and I replied to the person that I would "ask" in this notes 
conference. I got an answer, I'm happy with that...


			Thanks for the answer

				Bill
934.4CSC32::BUTTERWORTHGun Control is a steady hand.Tue Aug 22 1995 14:4011
    Re -1.
    
    I hope I didn;t sound like I was jumping you!! I really didn't mean it
    that way. I was playing devils advocate actually. What we can do is
    look at some different methods for the next release of ECO/MUP kit.
    I wouldn't hold my breath for a ECO/MUP kit becuase those kits don't
    get tested like a release does. A cahnge like this appears minor on the
    surface but has major implications as we've already talked about.
    
    REgards,
       Dan
934.530188::CLEMENCEFri Aug 25 1995 13:312
	No offense taken...